1. Volume 25, Issue 14, Pages 1852-1859 (July 2015)
Dissociable Effects of Serotonin and Dopamine on the Valuation of Harm in Moral Decision Making 
Molly J. Crockett, Jenifer Z. Siegel, Zeb Kurth-Nelson, Olga T. Ousdal, Giles Story, Carolyn Frieband, Johanna M. Grosse-Rueskamp, Peter Dayan, Raymond J. Dolan 
Current Biology 
Volume 25, Issue 14, Pages (July 2015)
DOI: /j.cub
Copyright © 2015 The Authors Terms and Conditions

2. Figure 1 Experimental Design
In each trial, deciders chose between less money and fewer shocks versus more money and more shocks.
(A–D) The money was always for the decider, but in half the trials, the shocks were allocated to the decider (A and C), and, in the other half, the shocks were allocated to the receiver (B and D). In all trials, if the decider failed to press a key within 6 s, the highlighted default (top) option was registered; if the decider pressed the key, the alternative (bottom) option was highlighted and registered instead. In half the trials, the alternative option contained more money and shocks than the default (A and B), so action resulted in greater harm and profit. In the other half, the alternative option contained less money and shocks than the default (C and D), so inaction resulted in greater harm and profit.
Current Biology  , DOI: ( /j.cub )
Copyright © 2015 The Authors Terms and Conditions

3. Figure 2
Effects of Citalopram and Levodopa on Harm Aversion and Hyperaltruism in Study 1 and Study 2
(A) In study 1, Citalopram significantly increased harm aversion for self (κself) and others (κother) but did not affect the difference in harm aversion between self and others (i.e., hyperaltruism).
(B) Sorted effect sizes of hyperaltruism (defined by taking the difference between κother and κself) across participants for the placebo and citalopram groups in study 1. Black bars indicate hyperaltruistic subjects, while white bars indicate selfish subjects.
(C) In study 2, Levodopa did not affect harm aversion for self or others, but significantly decreased the difference in harm aversion between self and others.
(D) Sorted effect sizes of hyperaltruism (κother − κself) across participants for the placebo and levodopa groups in study 2.
∗p < 0.05; ∗∗p < 0.01; n.s., not significant. Error bars represent SEM difference between κself and κother.
Current Biology  , DOI: ( /j.cub )
Copyright © 2015 The Authors Terms and Conditions

4. Figure 3
Predictions of Fitted Regression Models of the Interaction of Drug and Effective Dosage on Harm Aversion for Self and Others and Hyperaltruism
(A) Citalopram increased harm aversion for self and others relative to placebo, more strongly for subjects with lower body weight (who thus received a higher effective dose).
(B) Levodopa reduced hyperaltruism relative to placebo, more strongly for subjects with lower body weight (who thus received a higher effective dose).
Current Biology  , DOI: ( /j.cub )
Copyright © 2015 The Authors Terms and Conditions

5. Figure 4 Effects of Hyperaltruism and Drugs on Response Times
(A and B) Hyperaltruism (κother − κself) was positively correlated with slower decisions for others relative to self in study 1 (A) and study 2 (B), as measured by the weight of a general linear model regressor indicating whether the outcome was for other (βother).
(C) Levodopa reduced slowing associated with deciding for others, relative to self, but did not affect speeding related to value differences (|ΔV|).
(D) Citalopram reduced speeding related to value differences but did not affect slowing associated with deciding for others, relative to self.
∗p < Error bars represent SEM. a.u., arbitrary units. See also Table S3.
Current Biology  , DOI: ( /j.cub )
Copyright © 2015 The Authors Terms and Conditions